CoreML image input format

I trained a model with TensorFlow, converted the .mlmodel model through tfcoreml, and observed the input format in Xcode, as shown in the following figure. Excuse me, how can I convert the input image into the following input format?

Menu